For people who've done work on metabolic disease, this paper in PNAS may come as a surprise, although there was a similar warning in January of this year. Acetyl CoA-carboxylase 2 (ACC2) has been seen for some years as a target in that area. It produces malonyl CoA, which is a very important intermediate and signaling molecule in fatty acid metabolism (and other places as well). A number of drug companies have taken a crack at getting good chemical matter (I'm no stranger to it myself, actually). A lot of the interest was sparked by reports of the gene knockout mice, which seem to have healthy appetites but put on no weight. The underlying reason was thought to be that fatty acid oxidation had been turned up in their muscle and adipose tissue - and a new way to burn off excess lipids sounded like something that a lot of people with excess weight and/or dyslipidemia might be able to use. What's more, the ACC2 knockout mice also seemed to be protected from developing insulin resistance, the key metabolic problem in type II diabetes. An ACC2 inhibitor sounds like just the thing.
Well, this latest paper sows confusion all over that hypothesis. The authors report having made some selective ACC2 knockout mouse strains of their own. If the gene is inactivated only in muscle tissue, the animals show no differences at all in body weight, composition, or food intake compared to control mice. What's more, when they went back and inactivated ACC2 in the whole animal, they found the same no-effect result, whether the animals were fed on standard chow or a high-fat diet. The muscle tissue in both cases showed no sign of elevated fatty acid oxidation. The authors state drily that "The limited impact of Acc2 deletion on energy balance raises the possibility that selective pharmacological inhibition of Acc2 for the treatment of obesity may be ineffective."
Yes, yes, it does. There's always the possibility that some sort of compensating mechanism kicked in as the knockout animals developed, something that might not be available if you just stepped into an adult animal with an inhibiting drug. That's always the nagging doubt when you see no effect in a knockout mouse. But considering that those numerous earlier reports of knockout mice showed all kinds of interesting effects, you have to wonder just what the heck is going on here.
Well, the authors of the present paper are wondering the same thing, as are, no doubt, the authors of that January Cell Metabolism work. They saw no differences in their knockout animals, either, which started the rethinking of this whole area. (To add to the confusion, those authors reported seeing real differences in fatty acid oxidation in the muscle tissue of their animals, even though the big phenotypic changes couldn't be replicated). Phrases like "In stark contrast to previously published data. . ." make their appearance in this latest paper.
The authors do suggest one possible graceful way out. The original ACC2 knockout mice were produced somewhat differently, using a method that could have left production of a mutated ACC2 protein intact (without its catalytic domain). They suggest that this could possibly have some sort of dominant-negative effect. If there's some important protein-protein interaction that was wiped out in the latest work, but left intact in the original report, that might explain things - and if that's the case, then there still might be room for a small molecule inhibitor to work. But it's a long shot.
The earlier results originated from the lab of Salih Wakil at Baylor (who filed a patent on the animals), and he's still very much active in the area. One co-author, Gerry Shulman at Yale, actually spans both reports of ACC2 knockout mice - he was in on one of the Wakil papers, and on this one, too. His lab is very well known in diabetes and metabolic research, and while I'd very much like to hear his take on this whole affair, I doubt if we're going to see that in public.